I don't think anyone working in science believes any model they investigate is going to work without qualification for all time (for a limited time, with current ability to verify, sure; Newtonian gravity worked that way for a long time, too.)

Science approaches the underlying reality asymptotically.